Artificial IntelligenceHigh Priority (8/10)Usa

Microsoft Unveils Three Homegrown AI Models for Speech and Image Processing

Microsoft has released public preview versions of three in-house machine learning models focused on speech recognition, speech synthesis, and image generation, available through its Azure AI Foundry platform.

Key Points

  • Three new homegrown ML models for speech recognition, speech synthesis, and image generation
  • Available through Azure AI Foundry platform in public preview
  • Signals Microsoft's strategy to diversify beyond OpenAI dependency
  • Partnership with OpenAI extended to 2032

Full Details

Microsoft announced three new homegrown AI models on April 2, 2026, marking a significant expansion of its AI portfolio beyond the OpenAI partnership. The models cover speech recognition, speech synthesis, and image generation capabilities. They are now available through Azure AI Foundry (formerly Azure AI Studio), Microsoft's platform for developing AI agents and applications. Naomi Moneypenny, who leads the Azure AI Foundry Models product team, announced the models in a blog post. The move comes after Microsoft renegotiated its agreement with OpenAI, extending the partnership to at least 2032. Microsoft's Copilot now focuses on four areas: Copilot experience, Copilot platform, Microsoft 365 apps, and AI models. The company also confirmed that Mustafa Suleyman will continue to lead Microsoft's AI research efforts.

Why It Matters

Microsoft's push to develop proprietary AI models indicates a strategic hedging approach - maintaining the OpenAI partnership while building internal capabilities. This could reduce costs and provide more control over AI offerings, potentially reshaping the competitive landscape in enterprise AI services.

Sourcetheregister.com

Get stories like this delivered daily

AI-curated news, personalized to your interests. Zero noise.

Start 7-Day Free Trial →

More in Artificial Intelligence